Create app.py

app.py

@@ -0,0 +1,114 @@
+
+pip install gradio
+
+from google.colab import drive
+from sklearn.model_selection import train_test_split
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn import svm
+import pandas as pd
+import numpy as np
+import random
+import gradio as gr
+
+# Mount Google Drive to access your dataset
+drive.mount("/content/gdrive")
+
+# Define classes for reviews and review container
+class Review:
+    def __init__(self, text, score):
+        self.text = text
+        self.score = score
+        self.sentiment, self.probability = self.get_sentiment()
+
+    def get_sentiment(self):
+        if self.score <= 2:
+            return "NEGATIVE", 1.0 if self.score == 0 else 0.8
+        elif self.score == 3:
+            return "NEUTRAL", 0.5
+        else: #Score of 4 or 5
+            return "POSITIVE", 1.0 if self.score == 5 else 0.8
+
+class ReviewContainer:
+    def __init__(self, reviews):
+        self.reviews = reviews
+
+    def get_text(self):
+        return [x.text for x in self.reviews]
+
+    def get_sentiment(self):
+        return [x.sentiment for x in self.reviews]
+
+    def get_probability(self):
+        return [x.probability for x in self.reviews]
+
+    def evenly_distribute(self):
+        negative = list(filter(lambda x: x.sentiment == "NEGATIVE", self.reviews))
+        positive = list(filter(lambda x: x.sentiment == "POSITIVE", self.reviews))
+        positive_shrunk = positive[:len(negative)]
+        self.reviews = negative + positive_shrunk
+        random.shuffle(self.reviews)
+
+# Load the dataset
+df = pd.read_json('/content/gdrive/My Drive/Datasets/Books_subset3.json', lines=True)
+
+# Create Review objects
+reviews = [Review(x, y) for x, y in zip(df['review_body'], df['star_rating'])]
+
+# Split data into train and test sets
+train,test = train_test_split(reviews, test_size=0.25, random_state=42)
+
+# Initialize ReviewContainer objects
+train_container = ReviewContainer(train)
+test_container = ReviewContainer(test)
+
+# Evenly distribute the data
+train_container.evenly_distribute()
+train_x = train_container.get_text()
+train_y = train_container.get_sentiment()
+
+test_container.evenly_distribute()
+test_x = test_container.get_text()
+test_y = test_container.get_sentiment()
+
+# Vectorize the data
+vectorizer = TfidfVectorizer()
+train_x_vectors = vectorizer.fit_transform(train_x)
+test_x_vectors = vectorizer.transform(test_x)
+
+# Train SVM model
+clf_svm = svm.SVC(kernel='linear', probability=True)
+clf_svm.fit(train_x_vectors, train_y)
+
+# Define function for Gradio interface
+def predict_sentiment(text):
+    # Vectorize the input text
+    text_vector = vectorizer.transform([text])
+
+    # Predict sentiment
+    sentiment = clf_svm.predict(text_vector)[0]
+
+    # Get probabilities for each class
+    probabilities = clf_svm.predict_proba(text_vector)[0]
+
+    # Convert probabilities to percentages
+    percentages = [round(prob * 100, 2) for prob in probabilities]
+
+    # Choose the sentiment label based on the predicted class
+    if sentiment == "POSITIVE":
+        return f"Positive ({percentages[1]}%)"
+    elif sentiment == "NEUTRAL":
+        return f"Neutral ({percentages[2]}%)"
+    else:
+        return f"Negative ({percentages[0]}%)"
+
+# Create Gradio interface
+iface = gr.Interface(
+    fn=predict_sentiment,
+    inputs=gr.Textbox(placeholder="Enter Text", lines=10, label="Enter your text here:"),
+    outputs=gr.Textbox(label="Sentiment"),
+    title="Sentiment Analysis Developed by Group-12(Ankit, Akshat, Gautam, Pritish) with ♥ from RCC Institute of Information Technology.",
+    description="Enter text and predict sentiment"
+)
+
+# Launch the interface
+iface.launch(inline=False)